林昭安 Chao-An Lin

GPU accelerated fluid dynamics simulations

Our computational fluid dynamics (CFD) research focuses on numerical methods, GPU acceleration, and complex fluid interactions. We explore advanced techniques such as the immersed boundary method, the lattice Boltzmann method (LBM), and multigrid acceleration to enable large-scale simulations with enhanced efficiency. A key aspect of our work involves GPU-accelerated simulations, encompassing turbulent flows, high-density ratio two-phase flow, and mesh-refinement techniques. Our research on multigrid-accelerated solvers further improves computational efficiency for unsteady flow simulations. We have also studied non-Newtonian and polymeric fluids, including direct numerical simulations of FENE-P fluids and polymer-turbulence interactions, shedding light on drag reduction mechanisms. Overall, our research advances computational methodologies and broadens their applications in fluid dynamics, emphasizing GPU acceleration and numerical efficiency for large-scale simulations.